Online since 2002. Over 3300 puzzles, 2600 worldwide members, and 270,000 messages.

TwistyPuzzles.com Forum

It is currently Thu Jul 24, 2014 1:14 am

All times are UTC - 5 hours



Post new topic Reply to topic  [ 5 posts ] 
Author Message
 Post subject: Quantifying and improving the look ahead
PostPosted: Tue Sep 14, 2010 12:31 pm 
Offline
User avatar

Joined: Sun Feb 14, 2010 11:24 am
I came up with a new technique for targeted practice, and cut my average from 28 sec to 26 after ONE session, after weeks of trying to cut time with other methods. I also achieved 5 sub 20s in one day having only ever done two before. Felt I had to share.

Obviously there is more than one way to cut time, we can learn new algs, to go from perhaps a 3 look to a 2 look OLL or something similar. You can practice algs over and over to get individual ones faster. You can get a faster cube or use a different lubricant. You can improve your look ahead, the usual method being to slow solve and watch more carefully.

This method seeks to improve the transition from one step to the next by focussing on them separately.

Basically I will solve the cube in two steps and stop in between. Obviously this will result in a faster than usual time, the difference between the times represents the room for improvement with look ahead.

Example: Complete whole cube - average 28 sec.
Complete F2L average 18 sec + complete LL average 8 sec. = average 26 sec

Stopping and taking a look saves 2 seconds of solve time on average. Being able to do so during a complete solve would yield the same full solve average.

Now what I found from some analysis was that my transition from cross to first pair was REALLY TERRIBLE. And first to second pair was not great either

e.g. Complete whole cube - average 28 sec.

Complete cross average 3.5 sec + finsh cube average 20 sec = 23.5 sec!

Complete cross + first pair average 8 sec + finish cube average 18 sec = 26 sec.

What I found is that by stopping the solve and taking time to look ahead before continuing led to radically different full solve times, depending on where I stopped. Ther bigger the gap between that time and the real average, the worse the transition is at that point, and the more room for improvement there is.

By examining my process at the first pair I realised I was sitting there staring at the cube like an idiot for about 2-3 seconds doing nothing, and examining pretty much every pair trying to decide which to do first. Trying to find and place the first pair as quickly as possible, I got the average for Cross +pair 1 down to 6.5 sec quite easily, shaved a second off the rest of the process to make 23-24 total average over about 40 solves.

When I went back to full solving without stopping, the targeted work on look ahead and segmenting the solve enabled me to get around a 26 average for 30 solves or so, which I also repeated the next day. I also achieved sub-25 for 10 solves, which has been my goal for weeks now.

Repeated practice of mid solve inspection trains you to plan and solve that particular step more quickly, and is very easy to insert into a real solve.

Obviously the results are pretty standard, working on F2L properly makes you faster, and look ahead is good. The point of the method is to analyse exactly which parts of your look ahead need the most work, quantify and address them, and measure improvement accurately.

Hope this works for others too!

_________________
3x3 PB : 17.20; avg 5 : 24.86
PLLs learned : 21; OLLs learned : 12


Top
 Profile  
 
 Post subject: Re: Quantifying and improving the look ahead
PostPosted: Tue Sep 14, 2010 12:34 pm 
Offline
User avatar

Joined: Thu Sep 24, 2009 12:21 pm
Location: Chichester, England
So your method is look ahead, and trying to increase the chances of PLL skip? That's what I do. I have no idea how I get them times I do when I solve the cross on top, intuitive F2L, 2 Look OLL, and not even full PLL!

_________________
3x3x3 single: 5.73 seconds.
3x3x3 average of five: 8.92 seconds.
3x3x3 average of twelve: 9.77 seconds.

Buy the Curvy Copter Skewb, NovaMinx, and more here!


Top
 Profile  
 
 Post subject: Re: Quantifying and improving the look ahead
PostPosted: Tue Sep 14, 2010 12:42 pm 
Offline
User avatar

Joined: Sun Feb 14, 2010 11:24 am
Nothing to do with engineering a PLL skip, but you could do it with this method. The key point is that the method doesn't involve learning anything new, just doing what you do now more efficiently by analysing how you do it. If you have some fat to trim in your times this will help you find it and cut down the time you are wasting between steps. I haven't had any good improvements since learning PLL, until now.

_________________
3x3 PB : 17.20; avg 5 : 24.86
PLLs learned : 21; OLLs learned : 12


Top
 Profile  
 
 Post subject: Re: Quantifying and improving the look ahead
PostPosted: Tue Sep 14, 2010 12:47 pm 
Offline
User avatar

Joined: Thu Sep 24, 2009 12:21 pm
Location: Chichester, England
So just lookahead? :lol:

Whatever part of the solve you are doing, if you do it intuitively, over time you develop your own mental algorithms. So unless you pause to attempt to either make a skip, or make the next step easier, you might as well learn to lookahead whilst solving.

_________________
3x3x3 single: 5.73 seconds.
3x3x3 average of five: 8.92 seconds.
3x3x3 average of twelve: 9.77 seconds.

Buy the Curvy Copter Skewb, NovaMinx, and more here!


Top
 Profile  
 
 Post subject: Re: Quantifying and improving the look ahead
PostPosted: Tue Sep 14, 2010 5:32 pm 
Offline

Joined: Thu Feb 12, 2009 5:22 pm
Location: Massachusetts
Yes, it's "just lookahead". But he's done something I've never seen before. He's actually quantified what he can improve on. It's simple, and one of those things where you go "Why didn't I think of that?".

Bravo.

_________________
-Jonny G.

f
Will you join the Rebelution?
The teenage rebellion against low expectations.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 5 posts ] 

All times are UTC - 5 hours


Who is online

Users browsing this forum: No registered users and 9 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  

Forum powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group